3 At the time of this release (SXEmacs 22.1.15), SXEmacs has the
4 following idiosyncrasies:
9 ** User init file (C-h v user-init-file)
11 SXEmacs looks for user init files in `user-init-directory'. The
12 preferred directory is: ${XDG_CONFIG_HOME}/sxemacs but it can fall
13 back to the old ~/.sxemacs directory.
17 ${XDG_CONFIG_HOME}/sxemacs
18 ${HOME}/.config/sxemacs # if $XDG_CONFIG_HOME is not set
19 ${HOME}/.sxemacs # if other dirs don't exist
21 You can also force the use of ~/.sxemacs regardless of the existence
22 of the XDG dir/var by setting $SXE_USE_LEGACY environment variable to
25 If you're coming from XEmacs, symlinking your old ~/.xemacs directory
26 to a SXEmacs location should be enough to get you up and running:
28 $ ln -svfn ${HOME}/.xemacs ${XDG_CONFIG_HOME}/sxemacs
30 BTW, unlike XEmacs, SXEmacs doesn't attempt to "migrate" your old init
31 file or Gnu/Emacs .emacs file.
35 *** System-wide Packages (late-packages)
37 The default location that SXEmacs searches for packages is
38 `$prefix/share/sxemacs/'. The same as for the user-init-file, a
39 symlink is all you need to get up and running.
41 $ ln -svfn /usr/local/lib/xemacs /usr/local/share/sxemacs
43 *** User Packages in ${HOME} (early-packages)
45 For packages that you keep in your ${HOME}, the preferred location is:
46 ${XDG_DATA_HOME}/sxemacs. This is normally ${HOME}/.local/share/sxemacs,
47 and SXEmacs will use that if ${XDG_DATA_HOME} is not set.
49 These packages may also be located in ~/.sxemacs if that is where you
50 have your user-init-directory set to.
58 *** FFI is not included with your distro
60 Sadly, some Linux distributions (hello Fedora) don't ship a libffi
61 package, and their GCC does NOT include libffi or FFI headers either.
62 In this instance you have 2 options...
64 1) Get the standalone package of libffi at
65 <http://sourceware.org/libffi/>.
67 2) Compile your own GCC from source, making sure you enable the java
68 compiler. Enabling java in your GCC build is the only way to get
71 Obviously, option #1 there is the easiest and quickest path to
72 FFI-enabled SXEmacsen, and it is the option that we recommend.
74 Oh, and please nag your distro to have FFI included by default.
76 *** FFI is included in your GCC but you see missing header errors
78 Often libffi headers aren't completely installed. If you are getting
79 errors in effi.c that seem to be hinged from something like...
81 /usr/include/ffi.h:63:23: ffitarget.h: No such file or directory
83 You need to find `ffitarget.h' and put it in the same directory as
84 your `ffi.h'. Your libffi came with GCC, so you'll find it within
87 $ dirname $(gcc -print-libgcc-file-name)
88 /usr/lib/gcc-lib/i586-pc-linux-gnu/3.3.1
90 Using that example, ffitarget.h would be in...
92 /usr/lib/gcc-lib/i586-pc-linux-gnu/3.3.1/libffi/
94 Just copy or symlink the ffitarget.h there to /usr/include
96 *** FFI on SELinux enabled machines
98 If you are running with SELinux enabled and configure fails with
99 messages like the following in `config.log'...
101 error while loading shared libraries: /usr/local/lib/libffi.so.1:
102 cannot restore segment prot after reloc: Permission denied
104 You need to correct the default security context for `libffi.so'.
106 $ chcon -t textrel_shlib_t /usr/local/lib/libffi.so
110 The autoconf tests for PostgreSQL support have changed. SXEmacs'
111 configure script now uses `pg_config' to determine whether or not to
112 enable PostgreSQL. Because of this you may have to set you $PATH
113 environment to include the pgsql bin directory. It is normally
114 `/usr/local/pgsql/bin/'. Another popular directory on Solaris 9 is
115 `/opt/crw/postgresql/bin/'. Check with your site administrator.
117 Bash users can do it like this...
119 export PATH=/usr/local/pgsql/bin:$PATH
121 *** Solaris 9 with 64-bit PostgreSQL
123 There has also been a report that on Solaris 9 you may also need to
124 configure with `--with-cflags='-mcpu=ultrasparc -m64''. Apparently
125 GCC on Solaris 9 defaults to building 32-bit, so you lose if you have
128 ** 64-bit test suite failure
130 We have had a couple of reports of the test suite failing on 64-bit
131 systems. The error is like this (or similar)...
133 Testing /usr/src/sxemacs/modules/ase/ase-heap-tests.el...
134 Loading ase_heap v0.0.0 (SXEmacs module: ase-heap)
135 Loaded module ase_heap v0.0.0 (SXEmacs module: ase-heap)Fatal error: assertion failed, file alloc.c, line 298, block != (void*)0xCAFEBABEDEADBEEF
136 make[3]: *** [check-am] Aborted
138 At this point we are not too sure exactly what the issue is. It looks
139 like it might be a bug in the malloc or free code of the libc. We do
140 know that not all 64-bit systems are affected, so far, only Fedora
141 Core 7, and Gentoo on x86_64.
143 One user has reported that using `-O1' in CFLAGS prevents it.
145 But even with this test failure, SXEmacs still runs and operates
146 without incident. In fact, the failure can't be reproduced when
147 running the test suite interactively. With that in mind, it should be
148 safe to install if you see this failure.
150 We'll endeavour to get to the bottom of this one ASAP, if you think
151 you can help, let us know.
153 ** m4, libtool, autoconf, automake, and whatnot
155 SXEmacs tries to cope with any combination of versions of the above
156 programs. However, there is one lower bound, autoconf 2.60, and
157 unfortunately this has an impact on the other parts of the build
160 To cut it really short, here is the minimum known-to-work combination:
161 - autoconf 2.62, automake 1.9.6, libtool 1.5.22, m4 1.4.6
163 In general we support (as of April, 2010):
164 - autoconf >= 2.62, including current git versions
165 - automake 1.9.6, 1.10, 1.10a, 1.11.1, and current git versions
166 - libtool 1.5.N with N >= 22, libtool >= 2.1a (current CVS version)
167 - m4 1.4.M with M >= 6 plus current git versions
169 Note that many libtool packages shipped with the distros (OpenSuSE,
170 Debian, just to name two) are _broken_. Make sure you compile
171 your own libtool in case you want to rerun autogen.sh or bootstrap
172 the build chain, and double check that you use --enable-ltdl-install
175 If you are on a platform that has its own _non_gnu libtool (like OS/X
176 Leopard) add --program-prefix=g to your gnu libtool configure so it
177 installs as glibtool and doesn't clobber your other one.
179 Sometimes it helps just to copy over the libtool script manually:
180 cp -a $(type -p libtool) ${top_builddir}
182 *** ylwrap fails with sed errors
184 Some versions of the ylwrap script provided by autotools uses commas
185 as separators in sed commands. As such if your build path uses commas
186 the ylwrap will fail.
188 Sample message (where the build path was /Users/njsf/Projects/SXEmacs/nsx-up/,,mac):
190 /Users/njsf/Projects/SXEmacs/nsx-up/,,mac/lib-src/make-docfile --modname cl-loop -E cl-loop.doc.c ../../../modules/cl/cl-loop.c
191 /bin/sh ../../../ylwrap ../../../modules/cl/cl-loop-parser.y y.tab.c cl-loop-parser.c y.tab.h cl-loop-parser.h y.output
192 cl-loop-parser.output -- bison -y -d
193 sed: 1: "s,/Users/njsf/Projects/ ...": bad flag in substitute command: 'm'
194 sed: 1: "s,/Users/njsf/Projects/ ...": bad flag in substitute command: 'm'
196 The workaround is to use a path without commas in it.
199 *** Missing libltdl.la (Solaris 2.8)
201 We've had a report that missing libtool on Solaris 2.8 isn't detected
202 and so the included libtool still isn't used. If you see an error
203 about a missing libltdl.la all you need to do is configure SXEmacs
210 *** configure on FreeBSD, NetBSD, OpenBSD, etc.
212 Building SXEmacs on *BSD as far as we know requires the GNU Bourne
213 Again SHell (bash) versions 3 or 4.
215 bash is available for all tier 1 architectures as a binary package and
216 and for tier 2/3 as a port.
218 To run configure successfully...
220 CONFIG_SHELL=/path/to/bash $CONFIG_SHELL configure [option, ...]
222 *** configure on FreeBSD
224 Turning on the use of libssp and -fstack-protector from configure
225 ( --with-error-checking=stack ) will result in a broken build.
227 Do not, under any circumstances, add -fstack-protector to CFLAGS, even
228 independently of the stack error checking option.
230 ** bdwgc and gcc and code optimisation
232 There are some weird optimisation issues with the Boehm-Demers-Weiser
233 garbage collector (hereafter BDWGC) and the GCC C compilers of the 2 and
234 3 series. The build will crash like this:
236 Loading build-autoloads.el...
237 Loading loadup-el.el...
238 Loading loadup.el...make[3]: *** [auto-autoloads.el] Segmentation fault
240 make[3]: Leaving directory
242 The C backtrace will look like:
244 #0 0xbff9a2f0 in ?? ()
245 #1 0xb7eaf7d6 in GC_invoke_finalizers () at finalize.c:787
246 #2 0xb7eaf8ed in GC_notify_or_invoke_finalizers () at finalize.c:844
247 #3 0xb7eb2c8c in GC_generic_malloc (lb=32, k=0) at malloc.c:190
249 If this is true for you, you may want to try another optimisation level:
251 ./configure CFLAGS="-g -O2"
253 If this still does not work out either dispense with BDWGC support or
254 use a recent C compiler. ATTOW, all GCC 4.x compilers (including SVN)
259 ENT is basically a conglomerate of internally and externally implemented
260 arithmetics. Hence it supports a number of libraries, some of which
261 overlap in their functionality, some others do not but then break at the
264 One of the most likely problems is the GMP vs. MPFR issue. In past
265 times, mpfr (a multiprecision library for floats with exact rounding
266 facilities) has been a part of the GMP distribution. Later on, mpfr got
267 separated from it and has been developed independently while the version
268 of mpfr which ships with GMP stayed the same. Now that scenario is
271 Inattentive distributions (like Fedora) still deliver packages of GMP
272 with the old'n'incompatible mpfr library. SXEmacs will disable the MPFR
273 support on such systems by default (at configure time). However, if you
274 install a supported version of mpfr in parallel to the packaged ones on
275 such a system SXEmacs autodetection correctly reports that a sane
276 version of mpfr is available and enables it. Nonetheless, the according
277 build may fail (or the build may even succeed but calling the binary may
280 number-mpfr.o: In function `ent_lt_BIGFR_T':
281 /home/martin/src/edit/sxemacs-main/src/number-mpfr.c:661: undefined
282 reference to `mpfr_less_p'
283 number-mpfr.o: In function `ent_gt_BIGFR_T':
284 /home/martin/src/edit/sxemacs-main/src/number-mpfr.c:671: undefined
285 reference to `mpfr_greater_p'
288 Especially note that we _only_ support the standalone version of MPFR,
289 and not the one distributed with GMP.
294 Badger your distributor and demand separate packages for GMP and
297 Remove the GMP package and install your own build -- available at
298 http://swox.com/gmp -- afterwards install your own build of mpfr (the
299 one from http://www.mpfr.org)
301 Reconfigure and rebuild SXEmacs afterwards.
304 ** Build fails because of missing makeinfo
306 Install the GNU texinfo package on your system. You'll need at least
310 ** MacOS X warns of a crash during configure
312 This is normal, as one of the tests made during configure (for the
313 realpath call correctness) induces as crash.
315 If you are developing SXEmacs and will do lots of runs of configure
316 and that dialog annoys you, consider issuing:
318 # Disable crash reporting
319 sudo launchctl unload -w /System/Library/LaunchDaemons/com.apple.ReportCrash.Root.plist
320 # Redo last configure
321 ./config.status --recheck
322 # Enable crash reporting
323 sudo launchctl load -w /System/Library/LaunchDaemons/com.apple.ReportCrash.Root.plist
325 Another alternative (not recommended) is to launch
327 /Developer/Applications/Utilities/CrashReporterPrefs
329 and configure the mode to server, but you will loose notifications of
330 crashes on all applications.
332 In order to give SXEmacs developers with good diagnosis information it
333 is recommended the mode be Developer.
337 SXEmacs does build and run on OpenIndiana (151a) but you will need to
338 install a few files/packages beforehand. Namely...
340 Common Name OpenIndiana Package Name
344 automake automake-110
346 libtool libtool (also install libltdl)
353 Yes, you read that right... to get pkg-config you must install the
354 "gettext" package. :-)
356 In that list, `bison', `gmp', and `mpfr' are not critical, but you
357 will get extra functionality in your SXEmacs if you have them.
359 *** automake additional instructions for OpenIndiana
361 When you install the automake-110 OpenIndiana package it won't set up
362 the symlinks to /usr/bin/automake or /usr/bin/aclocal. Fix that
365 sudo ln -sv automake-1.10 /usr/bin/automake
366 sudo ln -sv aclocal-1.10 /usr/bin/aclocal
368 *** Running SXEmacs configure on OpenIndiana
370 There's one more quirk with OpenIndiana when you try to run SXEmacs'
371 configure... you MUST set $CONFIG_SHELL
373 CONFIG_SHELL=/bin/bash ../configure [opts]
376 ** make does not stop on subdirectory build failure
378 Due to a bug in the make argument parsing in code generated by
379 autoconf it is possible for make not to stop when a subdirectory fails.
381 This failure occurs for instance when the make command line has a variable
382 assignment which has a value with a - and k. Example:
384 make CFLAGS="-Wall -fpacked -fpedantic" build-report
390 We have identified 2 packages so far that don't work "out of the box"
391 with SXEmacs. In both of these the problem is with parsing version
392 information. Patches have been sent to the appropriate maintainer to
393 fix the problem and are included here in case the packages haven't
394 been updated by the time you install SXEmacs.
396 Update: The EFS, and Dired XEmacs packages that are currently
397 available from the "Pre-Releases" area of XEmacs package mirrors are
398 both now compatible with SXEmacs and do not need the patches mentioned
401 * Problems with running SXEmacs
402 ===============================
406 *** ffi-wand.el refuses to load.
408 Can't load library `libMagickWand': libgomp.so.1: shared object cannot be
411 If you get that error when trying to load ffi-wand, it is because you
412 have a ImageMagick that is using OpenMP (currently only svn HEAD). To
413 fix this you will need to rebuild ImageMagick, making sure that you
414 configure it using --disable-openmp.
416 See: <http://issues.sxemacs.org/show_bug.cgi?id=104>
418 ** Multimedia Goodness
420 *** SXEmacs hangs or crashes during (init-asynchronousity).
422 This is most likely a known effect (we do not want to call it bug,
423 since there is no definite location) with certain (g)libc and kernel
424 combinations under Linux. If it crashes analyse the core file, it
425 should look like this:
427 #0 0x4014ebc4 in __sigsuspend (set=0xbffffbb4) at
428 ../sysdeps/unix/sysv/linux/sigsuspend.c:48
429 #1 0x40101b34 in __pthread_wait_for_restart_signal (self=0x401116e0) at
431 #2 0x40101138 in __pthread_create_2_1 (thread=0x206f8dc, attr=0xbffffc58,
432 start_routine=0x20043ac <console>, arg=0xbffffd88) at restart.h:26
434 A definite fault-prone setup is using kernel 2.6.x in conjunction with
437 *** SXEmacs hangs or crashes before it ought to playback sound.
439 As before, this is most likely a suspicious (g)libc/kernel
442 *** SXEmacs dumps core when using the ALSA audio device
444 This has been reported to happen with old ALSA libraries (1.0.3 to be
445 precise). At the moment it is uncertain at which version these
446 problems disappear (no developer wants to downgrade to a non-working
447 ALSA :D). We highly suggest to use the version 1.0.10 and above, or
450 *** SXEmacs in async mode does not play simultaneous sounds with ALSA
452 This is due to missing (hardware-)mixing capabilities of your
453 soundcard. There is a user-space plugin called dmix, which can
454 effectively circumvent this issue.
456 *** SXEmacs crashes when using state sentinels with asynchronous sounds
458 This is a known bug (#13 in our bug database). At the moment the only
459 advise we can give is: do not use sentinels before 22.1.7.
460 Also see our bug database at http://issues.sxemacs.org
462 *** make-media-stream seems to recognise any file as valid audio
464 This is a known issue with fully-featured ffmpeg builds. The current
465 code in SXEmacs blindly relies on FFmpeg when it reports a file or
466 string as valid audio. There is no way to double-check that at the
467 moment. However, you can perform the additional check yourself if
468 you have taglib installed. Use the included ffi-taglib.el.
471 *** XCreateIC fails at startup
473 SXEmacs sometimes fails to create the input context with XCreateIC on
474 non-C languages. SXEmacs will include the values of the LANG and
475 XMODIFIERS environment variables which influence the behavior of
476 XCreateIC. Failures have been observed with XMODIFIERS=@im=ibus
479 * Original XEmacs PROBLEMS File
480 ===============================
482 The original XEmacs PROBLEMS file may be found in the SXEmacs
483 source distribution as PROBLEMS.XEmacs - while many issues mentioned
484 have since been fixed, it is preserved for posterity.